PROJECT SUMMARY It has long been argued that the brain is most plastic during early development and that after brain injury, children reorganize cortical functions to healthy areas; but there is little understanding of these processes. Many investigators have shown that, when the left hemisphere (LH) is damaged, language can be subserved by the right hemisphere (RH). Here we ask how the linguistic functions of LH and RH are organized after LH injury and how well this atypical organization supports linguistic abilities. We also investigate the complementary questions regarding how these functions are organized after RH injury. We will study the long-term outcomes after ischemic perinatal stroke (IPS), due to its relatively clean properties for scientific analysis. IPS occurs in 1/4000 births and is the leading known cause of cerebral palsy. Many children with LH perinatal stroke develop near-normal language and cognitive abilities, but a substantial number show later language and cognitive impairments. Most literature shows that, after early LH stroke, language `reorganizes' to homotopic RH areas. Our pilot data suggest that, under these circumstances, typical LH and RH language functions divide up the RH cortical territory in new ways. We hypothesize that these functions develop best when different language functions find distinct RH localizations; overlap of linguistic functions will lead to more impaired performance. We will test this hypothesis by investigating language after LH perinatal injuries. We will also ask the same questions after RH perinatal injuries. We will recruit 30 teenagers and young adults who have had a LH perinatal stroke involving the middle cerebral artery and 30 who have had a similar RH perinatal stroke, compared with 30 healthy controls matched in age and SES. While IPS is rare, our research team is ideal for successful recruiting. We have developed our tasks and tested 16 children with perinatal strokes. We will administer a battery of behavioral and neuropsychological tests and 5 fMRI tasks. Two fMRI tasks (examining naming and sentence processing) test language functions that activate LH language areas in healthy controls. Two fMRI tasks (examining vocal emotion and intonation) test functions ordinarily controlled by the RH. Together these tasks examine how language abilities are affected when they develop in one hemisphere. An additional fMRI task examines specific object recognition (word recognition, ordinarily in LH VWFA, and face recognition, ordinarily in RH FFA) to ask if these functions are altered when their typical neural substrates are intact but language is shifted. Our behavioral tests assess how well all of these skills operate when neural organization is atypical. These studies will contribute to our understanding of the flexibility and limitations of early brain plasticity. In the future our findings can also help to provide a foundation for developing techniques to stimulate the most successful types of functional organization after early brain injuries.